1 running head: fewer crown roots improve n capture in

1

Running head: Fewer crown roots improve N capture in maize 1

Corresponding author: 2

Jonathan Paul Lynch, Department of Plant Science, The Pennsylvania State University, 3

University Park, PA 16802, USA, Telephone number: 814-8632256, [email protected] 4

Journal research area: Ecophysiology and Sustainability 5

Plant Physiology Preview. Published on April 4, 2014, as DOI:10.1104/pp.113.232603

Copyright 2014 by the American Society of Plant Biologists

www.plant.org on April 12, 2014 - Published by www.plantphysiol.orgDownloaded from Copyright © 2014 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org/


http://www.plant.org


2

Low crown root number enhances nitrogen acquisition from low nitrogen soils in 6

maize (Zea mays L.). 7

Patompong Saengwilai1, Xiaoli Tian2,3, and Jonathan Paul Lynch1,2 8

Summary: low crown root number improves nitrogen acquisition in maize by enhancing 9

deep soil exploration in low N soils. 10






3

This research was supported by the Howard G. Buffett Foundation and the National 11

Science Foundation- PGRP grant DBI 0820624. 12

1Intercollege Program in Plant Biology, The Pennsylvania State University, University 13

Park, PA 16802, USA 14 2Department of Plant Science, The Pennsylvania State University, University Park, PA 15

16802, USA 16 3State Key Laboratory of Plant Physiology and Biochemistry, and Department of 17

Agronomy, China Agricultural University, Beijing 100193, China 18

For correspondence: E-mail [email protected] 19






4

Abstract 20

In developing nations, low soil nitrogen (N) availability is a primary limitation to crop 21

production and food security, while in rich nations, intensive N fertilization is a primary 22

economic, energy, and environmental cost to crop production. It has been proposed that 23

genetic variation for root architectural and anatomical traits enhancing exploitation of 24

deep soil strata could be deployed to develop crops with greater N acquisition. Here we 25

provide evidence that maize (Zea mays L.) genotypes with few crown roots (crown root 26

number: CN) have greater N acquisition from low N soils. Maize genotypes differed in 27

their CN response to N limitation in greenhouse mesocosms and in the field. Low CN 28

genotypes had 45% greater rooting depth in low N soils than high CN genotypes. Deep 29

injection of 15N-labeled nitrate showed that low CN genotypes acquired more N from 30

deep soil strata than high CN genotypes, resulting in greater photosynthesis and total 31

nitrogen content. Under low N, low CN genotypes had greater biomass than high CN 32

genotypes at flowering (85% in the field study in the US and 25% in South Africa). In the 33

field in the US, 1.8x variation in CN was associated with 1.8x variation in yield reduction 34

by N limitation. To our knowledge, this is the first report of the utility of CN for nutrient 35

acquisition. Our results indicate that CN deserves consideration as a potential trait for 36

genetic improvement of nitrogen acquisition from low N soils. 37

Keywords: Zea mays L., crown root number, CN, mesocosm, nitrogen, 15N 38






5

Introduction 39

Maize (Zea mays L.) is one of the world’s most important crops and is a staple food in 40

Latin America and Africa. Maize production requires a large amount of fertilizer, 41

especially nitrogen. In the USA, N fertilizers represent the greatest economic and energy 42

costs for maize production (Ribaudo et al., 2011). However, on-farm studies across the 43

North-central USA revealed that more than half of applied N is not taken up by maize 44

plants and is vulnerable to losses from volatilization, denitrification, and leaching, which 45

pollute air and water resources (Cassman, 2002). Conversely, in developing countries 46

suboptimal nitrogen availability is a primary limitation to crop yields and therefore food 47

security (Azeez et al., 2006). Increasing yield in these areas is an urgent concern since 48

chemical fertilizers are not affordable (Worku et al., 2007). Cultivars with greater 49

nitrogen acquisition from low N soils could help alleviate food insecurity in poor nations 50

as well as reduce environmental degradation from excessive fertilizer use in developed 51

countries. 52

The two major soil N forms available to plants are ammonium and nitrate. Nitrate is the 53

main N form in most maize production environments (Miller and Cramer, 2004). Nitrate 54

is highly mobile in soil and the spatiotemporal availability of soil N is rather complex. In 55

the simplest case nitrogen fertilizers applied to the soil and/or nitrogen released from 56

mineralization of soil organic matter are rapidly converted to nitrate by soil microbes. 57

After irrigation and precipitation events, nitrate moves with water to deeper soil strata. 58

Leaching of nitrate from the root zone has been shown to be a significant cause of low 59

recovery of N fertilizer in commercial agricultural systems (Cassman et al., 2002; Raun 60

& Johnson, 1999). Differences in root depth influence the ability of plants to acquire N. 61

Studies using 15Nitrogen (15N) labeled nitrate placed at different soil depths showed that 62

only plants with deep rooting can acquire N sources from deep soil strata, which would 63

otherwise have been lost through leaching (Kristensen & Thorup-Kristensen, 2000; 64

Kristensen & Thorup-Kristensen, 2004). Therefore selection for root traits enhancing 65

rapid deep soil exploration could be used as a strategy to improve crop N efficiency. 66

The maize root system consists of embryonic and post-embryonic components. The 67

embryonic root system consists of two distinct root classes: a primary root and a variable 68






6

number of seminal roots formed at the scutellar node. The post-embryonic root system 69

consists of roots that are formed at consecutive shoot nodes and lateral roots, which are 70

initiated in the pericycle of all root classes. Shoot-borne or nodal roots that are formed 71

below ground are called “crown roots” whereas those that are formed above ground are 72

designated “brace roots” (Hochholdinger, 2009). While the primary root and seminal 73

roots are essential for the establishment of seedlings after germination, nodal roots and 74

particularly crown roots make up most of the maize root system and are primarily 75

responsible for soil resource acquisition later in development (Hoppe et al., 1986). 76

Lynch (2013) proposed an ideotype for superior N and water acquisition in maize called 77

“Steep, Cheap and Deep (SCD)”, which integrates root architectural, anatomical, and 78

physiological traits to increase rooting depth and therefore the capture of N in leaching 79

environments. One such trait is crown root number (CN). CN is an aggregate trait 80

consisting of the number of belowground nodal whorls and the number of roots per 81

whorl. The crown root system dominates resource acquisition during vegetative growth 82

after the first few weeks and remains important during reproductive development 83

(Hochholdinger et al., 2004). CN in maize ranges from 5 to 50 under fertile conditions 84

(Trachsel et al., 2011). At the low end of this range, crown roots may be too spatially 85

dispersed to sufficiently explore the soil. There is also a risk of root loss to herbivores 86

and pathogens. If roots are lost in low N plants, there may be too few crown roots left to 87

support the nutrient, water, and anchorage needs of the plant. At the high end, a large 88

number of crown roots may compete with each other for water and nutrients as well as 89

incur considerable metabolic costs for the plant (Fig 1). The SCD ideotype proposes that 90

there is an optimal number of crown roots (CN) for N capture in maize (Lynch, 2013). 91

Under low N conditions, resources for root growth and maintenance are limiting, and 92

nitrate is a mobile resource that can be captured by a dispersed root system. Optimal CN 93

should tend toward the low end of the phenotypic variation to make resources available 94

for development of longer, deeper roots rather than more crown roots. According to the 95

SCD ideotype, in low N soils, maize genotypes with fewer crown roots could explore 96

soils at greater depth resulting in greater nitrogen acquisition, growth, and yield than 97

genotypes with many crown roots. 98






7

The objective of this study was to test the hypotheses that: (i) low CN genotypes have 99

greater rooting depth than high CN genotypes in low N soils; (ii) low CN genotypes are 100

better at acquiring deep soil N than high CN genotypes; (iii) low CN genotypes have 101

greater biomass and yield than high CN genotypes in low N conditions. 102

Results 103

N stress effects on CN 104

In mesocosms, nitrogen limitation reduced crown root number by 26% (p<0.001) at 28 105

days after planting (DAP). The CN ranged from 3 to 9 under low N conditions. The six 106

genotypes responded differently to N limitation. OHW3, OHW74, OHW 61, and 107

IBM133 showed significant reduction in CN whereas OHW 170 and IBM 123 108

maintained their CN under low N conditions (Fig 2). Nitrogen limitation reduced the 109

average crown root whorl number from 2.75 to 2.13 (p<0.05; Fig 3A). Nitrogen 110

limitation did not affect the number of roots in the first whorl but significantly reduced 111

the number of roots of the second, third, and forth whorl, particularly low CN genotypes 112

(Fig 3B, supplemental Fig S1). 113

At the field site in the USA (US2011), N limitation reduced CN by 21% at flowering. 114

The CN ranged from 24 to 44 under low N conditions. The genotypes responded 115

differently to N limitation. Nitrogen limitation reduced CN in genotypes NYH76, 116

NYH57, and NYH212, but did not significantly affect CN in the three IBM lines (Fig 117

4A). At the field site in South Africa in 2011 (SA2011), the CN ranged from 21.5 to 35.5 118

under low N conditions. The six genotypes were grouped as high or low CN based on the 119

mean difference in CN under low N conditions. The high CN genotypes consisted of 120

IBM123, OHW3, and OHW170. The low CN genotypes consisted of IBM133, OHW61, 121

and OHW74. Means comparison showed that no genotype had a significant decrease of 122

CN under N limitation (Fig 4B), but ANOVA grouping genotypes into the two categories 123

of high CN or low CN showed a significant reduction of CN by N limitation (p<0.05), 124

with high CN genotypes having 10 more crown roots than low CN genotypes under low 125

N conditions. A different set of genotypes was planted at the field site in South Africa in 126

2012 (SA2012). In 2012 the CN ranged from 30 to 46.5 under low N conditions. There 127






8

was no significant effect of N stress on the average CN of these genotypes. Nitrogen 128

limitation affected CN in only one genotype, IBM165, and in this instance actually 129

increased CN (Fig 4C). 130

CN effects on rooting depth and N acquisition 131

In mesocosms, the genotypes were grouped into high CN and low CN genotypes based 132

on the average value of CN. The high CN genotypes consisted of OHW 170, OHW3, and 133

IBM133; the low CN genotypes consisted of OHW61, OHW74, and IBM123. We found 134

that most low CN genotypes had greater rooting depth than high CN genotypes under low 135

N conditions (Fig 5A; p<0.05). Primary roots, seminal roots, and crown roots of low CN 136

genotypes had greater rooting depth (p<0.05) than those of high CN genotypes (Fig 5B). 137

In SA2011 N limitation slightly increased maximum rooting depth (D95) from 30.5 to 138

37.2 cm but the effect was not significant. Low CN genotypes had significantly greater 139

rooting depth than high CN genotypes (Fig 5C) under low N conditions. The low CN 140

genotypes had a D95 value of 34.4 cm whereas for high CN genotypes the D95 value was 141

26.7 cm (p<0.05, Fig 5C). In UA2011 and SA2012 Low CN genotypes again had 142

significantly greater rooting depth than high CN genotypes (Fig 6A, supplemental Fig 143

S2). To investigate whether low CN genotypes were better at acquiring N from deep soil 144

strata, we injected 15N-labelled nitrate in the soil at a depth of 50 cm at SA2012. One 145

week after the 15N application we found that low CN genotypes had greater 15N content 146

in shoot tissues than high CN genotypes under low N conditions (Fig 6B). 147

CN effects on plant growth and yield 148

In mesocosms N limitation reduced shoot mass by an average of 45%. Shoot biomass and 149

leaf photosynthetic rate were affected by CN (Table I, II, supplemental table S1). 150

ANCOVA and correlation analyses showed that under low N conditions, plants with low 151

CN had greater leaf photosynthetic rates, canopy photosynthetic rates, tissue N content, 152

and shoot mass, than plants with high CN (Table I,II). There was no significant 153

relationship between these variables and CN under high N conditions (data not shown). 154






9

In the field trials N limitation reduced shoot mass by an average of 20% in SA2011 and 155

by 24% in SA2012. ANCOVA and correlation analyses showed that under low N 156

conditions, low CN genotypes had greater leaf photosynthetic rates, tissue N content, and 157

shoot dry weight than plants with high CN at SA2011 (Table I,III, supplemental table 158

S2). There was no significant relationship between these variables and CN under high N 159

conditions (data not shown). 160

In US2011 N limitation reduced shoot mass by 34% at flowering (8 weeks after 161

planting). Grain yield was reduced by 39% in low N soils. ANCOVA and correlation 162

analyses showed that under low N conditions, low CN genotypes had greater tissue 163

nitrogen content and shoot dry weight than high CN genotypes (Table I,III, supplemental 164

table S2). Low CN genotypes had greater percent grain yield than high CN genotypes 165

under low N conditions (Fig 7). Genotypic variation of 1.8x in CN was associated with 166

1.8x variation in yield reduction by N limitation (Fig 7). 167

Discussion 168

We demonstrate that low crown root number (CN) improves nitrogen acquisition by 169

enhancing deep soil exploration in low N soils. Genotypes differed in their CN response 170

to N limitation (Figs 2,4). Maize lines with low CN had greater rooting depth than high 171

CN genotypes (Figs 5,6) and acquired more 15N labeled nitrate applied in deep soil in the 172

field (Fig 6). Low CN genotypes had greater tissue nitrogen content and shoot biomass 173

than high CN genotypes under low N conditions in all environments tested (Fig 6,Table 174

I). Finally, low CN genotypes had greater percent grain yield than high CN genotypes in 175

the field under low N conditions (Fig 7). 176

This study is focused on the physiological utility of CN for N acquisition in low N 177

environments. The use of monogenic mutantsis not suitable for this study, since CN is a 178

quantitative trait controlled by several alleles in unknown ways (Burton, 2010). To date 179

genes controlling the development of root architecture such as RTCS and RL have been 180

identified (Jenkins, 1930; Hetz et al., 1996; Hochholdinger et al., 2004). However 181

mutations in these genes affect the development of other root classes (rtcs) and plant 182

vigor (rl) and thus are not desirable for our purpose. In this study, we selected near 183






10

isophenic lines from maize recombinant inbred lines (RILs) that vary in CN but are 184

similar in other phenotypic traits such as root angle and branching. RILs are suitable for 185

this study because they are closely related genotypes with highly similar genetic 186

backgrounds, thereby minimizing the risk of effects from genetic interactions, epistasis, 187

and pleiotropy, which may confound the interpretation of results from comparisons of 188

unrelated lines (Zhu et al., 2005; Zhu et al., 2006). In addition, each experiment consisted 189

of RILs from different populations representing high and low CN. The fact that our 190

results were consistant among different expriments with different set of RILs indicates 191

that the utility of CN for N capture is independent of the specific genotypic context. 192

In the greenhouse we used mesocosms to create nitrogen leaching environments 193

comparable to conditions in well-drained agricultural soils. The mesocosms also permit a 194

detailed investigation of root distribution by depth since entire root systems can be 195

excavated. Gaudin et al. (2011) reported that maize responded to N limitation by 196

increasing the length of individual crown roots while reducing CN (Gaudin et al., 2011). 197

These results are consistent with those of Tian et al (2008), who demonstrated that high 198

nitrate inhibits maize root elongation and is accompanied by decreasing IAA levels in the 199

roots (Tian et al., 2008). In our study, we found that not all maize genotypes reduced CN 200

in response to N limitation. For example, genotypes such as IBM133, OHW3, OHW61, 201

and OHW74 significantly reduced CN in the mesocosms under low N conditions but 202

maintained their CN in the field (Fig 2,4). These results indicate that CN response to N 203

limitation depends on genotypes and environments. In the mesocosms where CN was 204

significantly reduced by N limitation, we found that reduced CN was attributable to 205

decreased crown root whorl number and decreased number of roots per whorl (Fig 3A, 206

3B). Nitrogen stress did not affect the number of roots of the first whorl, which is the 207

earliest to emerge from the stem node, suggesting that plants may exhaust seed N 208

reserves prior to or during the development of the second whorl crown roots. 209

We found that high CN genotypes had shallower primary, seminal, and crown roots than 210

low CN genotypes under low N conditions (Fig 5). This result supports the hypothesis 211

that there exist tradeoffs between the number of crown roots and growth of different root 212

classes. These results are consistent with reports in other crop species. In wheat and 213






11

barley, the removal of nodal roots stimulates the growth and activity of the seminal roots 214

(Krassovsky, 1926). In common bean, increased carbon allocation to adventitious roots 215

was related to decreased allocation to tap and basal roots, which affected total root length, 216

soil exploration, and P acquisition under suboptimal P conditions (Walk et al., 2006), and 217

removal of a specific root class led to an increase in the relative proportion of the 218

remaining root classes (Rubio and Lynch, 2007). In maize the majority of axial roots in 219

the root system are crown roots. The diameter of crown roots of the third whorl and 220

subsequent nodes are much larger than that of the primary and seminal roots, and these 221

roots are thus a greater sink for photosynthates. High CN genotypes must maintain the 222

growth and development of many crown roots, which would constrain the growth and 223

elongation of crown roots and other root classes, resulting in shallower root systems 224

compared to those of low CN genotypes (Fig 5,6). In addition, competition among roots 225

within the root system for soil resources is greater in high CN genotypes, especially for a 226

mobile resource like nitrate. The effect of reduced CN on soil exploration and N 227

acquisition could result from reduced root competition for internal and external resources, 228

as proposed by Lynch (2013). 229

We investigated the ability of low CN genotypes to take up N from deep soil layers in the 230

field in SA2012 by injection of 15N-labelled nitrate in the soil at 50 cm depth within a 231

planting row adjacent to the plants. We found that low CN genotypes had greater 15N 232

uptake than high CN genotypes (Fig 6B). Soil nitrate analysis showed that nitrate was 233

indeed more abundant in deep soil layers than in topsoil at the time of harvest (data not 234

shown), thus, deep-rooting low CN genotypes are able to acquire N deep in the soil 235

profile better than high CN genotypes. The ability to explore soils at greater depth and 236

acquire N from N source in deep soils means that low CN plants have greater usage of N 237

and thus have better N efficiency than high CN genotypes. Low CN plants could also 238

reduce N leaching, thereby reducing environmental pollution. 239

Photosynthesis directly influences growth and yield of crop plants (Gastal and Lemaire, 240

2002). The rate of photosynthesis depends on content of N in the leaf tissue because 241

photosynthetic proteins, including Rubisco and light harvesting complex proteins, 242

represent a large proportion of total leaf N (Evans, 1983). We found that low CN 243






12

genotypes had greater tissue N content, which resulted in greater photosynthetic rates, 244

and shoot biomass than high CN genotypes in greenhouse and field studies (Table 245

I,II,III). In US2011, 1.8x genotypic variation in CN was associated with 1.8x variation in 246

yield loss due to N limitation (Fig 7). This is important especially for developing 247

countries where yield of maize is less than 10% of its yield potential (Lynch, 2007). 248

Considering the range of reported CN in field-grown plants of 5-50 (Trachsel et al., 2010) 249

and 10-32 (Bayuelo-Jiménez et al., 2011), our range of CN (20-45) falls between the 250

medium to high range of phenotypic variation observed in maize. We propose that in 251

extremely low CN phenotypes, roots may be too spatially dispersed to sufficiently 252

acquire soil resources and such plants may be susceptible to lodging (Hetz et al., 1996). 253

Additionally, plants with very low CN may be at risk of root loss due to herbivores and 254

pathogens. This is particularly important for low-input agroecosystems where root 255

survivorship is low. In this case the optimum number of CN would be large enough to 256

allow rapid recovery from root damage but not too large to compete for internal and 257

external resources. The optimum range of CN is likely to be dependent upon soil type and 258

the severity of biotic and abiotic stresses. We anticipate that the optimum range of CN is 259

also at the low end of the range of variation under drought and is likely to be greater in 260

low density plantings, in fine-textured soils with slow leaching, and in soils with 261

suboptimal availability of immobile nutrients such as phosphorus (P) and potassium (K), 262

which are abundant in the topsoil. Greater CN may be beneficial to plants in low-input 263

systems in which N continues to be available in the topsoil as a result of mineralization of 264

organic matter (Poudel et al., 2001). However, many low-input systems are subject to 265

drought in addition to suboptimal N availability. In this case, low CN enhancing deep soil 266

exploration may be preferable to high CN since low CN supports deep root system so the 267

shallow portion of deep roots can acquire shallow N resources while the deep portion can 268

explore deep soil for water resources. 269

Functional-structural modeling could be helpful in identifying optimum CN for specific 270

environments as well as studying interactions between CN and other root traits. Recently, 271

York et al. (2013) used the functional-structural plant model SimRoot, to observe 272

interactions between CN and root cortical aerenchyma (RCA). They found that the 273






13

synergistic effects of CN and RCA on plant growth were greater than the additive effects 274

by 32% at medium N and by 132% at medium P (York et al., 2013). In addition, an 275

optimum number of crown roots can also interact with other traits enhancing deep soil 276

exploration, such as steep root growth angle and few but long root branches, and may 277

synergistically enhance resource acquisition under drought and suboptimal availability of 278

mobile nutrients (Lynch, 2013). 279

The concept of optimum CN enhancing root growth and soil exploration under water and 280

nutrient limiting conditions supports the rhizoeconomic paradigm, which considers the 281

benefits and the costs of root traits as direct metabolic costs and as trade-offs and risks 282

(Lynch and Ho, 2005; Nord and Lynch, 2009). We suggest that the optimum CN concept 283

can be applied to other crop species in which nodal roots represent a major portion of the 284

root system such as rice (Oryza sativa), wheat (Triticum aestivum L.) and barley 285

(Hordeum vulgare L.) (Krassovsky, 1926; de Dorlodot et al., 2007; Coudert et al., 2010). 286

Our results are entirely supportive of the CN component of the SCD ideotype (Lynch 287

2013). The SCD ideotype applies to both water and N capture, since both of these soil 288

resources are often localized in deep soil strata under limiting conditions. The fact that 289

CN affects rooting depth and therefore N capture suggests that this trait should also be 290

useful for water capture from drying soil, especially in terminal drought scenarios (Lynch 291

2013). 292

Genotypic differences in crown root number have been reported in several crop species 293

including maize and its relatives within Zea (Bayuelo-Jiménez et al., 2011; Burton et al., 294

2013; Lynch, 2013; Trachsel et al., 2010). Moreover, CN is a heritable trait (Jenkins, 295

1930) and genes affecting CN expression have been identified (Jenkins, 1930; Hetz et al., 296

1996; Taramino et al., 2007) making CN a feasible target for plant breeding. To our 297

knowledge, this is the first report of the utility of CN for improving nutrient acquisition. 298

Our results support the hypothesis that CN affects rooting depth and soil N acquisition, 299

and thus merits investigation as a potential element of more N-efficient cultivars. 300






14

Materials and Methods 301

Greenhouse mesocosm study 302

Plant materials 303

Based on the results of screening experiments in mesocosms in the USA and in the field 304

in South Africa, recombinant Inbred Lines (RILs) IBM123 and IBM133 from the 305

intermated B73 and Mo17 (IBM) population (Lee et al., 2002; Sharopova et al., 2002) 306

and OHW3, OHW61, OHW74, and OHW170 from the cross between OH43 and W64a 307

(OHW) contrasting in crown root number were selected for this study. 308

Experimental design 309

The greenhouse experiment was a randomized complete block design. The factors were 310

two nitrogen regimes (high and low nitrogen conditions), six RILs, and four replicates. 311

Planting was staggered one week between replicates with time of planting as a block 312

effect. 313

Growth conditions 314

Plants were grown during October 13 to December 8, 2010 in a greenhouse located on 315

the campus of The Pennsylvania State University in University Park, PA, USA (40°48′N, 316

77°51′W), with a photoperiod of 14/10 h at 28/24 oC (light/darkness). Seeds were soaked 317

for 1 h in a fungicide solution containing benomyl (Benlate fungicide, E.I. DuPont and 318

Company, Wilmington, DE, USA) and 1.3 M metalaxyl (Allegiance fungicide, Bayer 319

CropScience, Monheim am Rhein, Germany) and then were surface-sterilized in 10% 320

NaOCl for 1 min. The seeds were pre-germinated in rolled germination paper (Anchor 321

Paper Company, St. Paul, MN, USA) soaked with 0.5 mM CaSO4 and placed in darkness 322

at 28oC in a germination chamber for two days. At planting, the plants were transferred to 323

mesocosms consisting of PVC cylinders 15.7 cm in diameter and 160 cm in height. The 324

mesocosms were lined with transparent high-density polyethylene film to facilitate root 325

sampling at harvest. The growth medium consisted of a mixture (volume based) of 50% 326

medium size (0.3 to 0.5 mm) commercial grade sand (Quikrete Companies Inc., 327

Harrisburg, PA, USA), 35% horticultural vermiculite, 5% Perlite (Whittemore 328

Companies Inc., Lawrence, MA, USA) and 10% topsoil. The topsoil was collected from 329






15

the Russell E. Larson Agricultural Research Center in Rock Springs, PA (Fine, mixed, 330

semiactive, mesic Typic Hapludalf, pH ≈ 6.7, silt loam). Thirty-three liters of the mixture 331

were used in each mesocosm to ensure the same bulk density of the medium. One day 332

before planting, the mesocosms were saturated with 5 liters of a nutrient solution adjusted 333

to pH 6. The nutrient solution for the high N treatment consisted of (in µM): NO3 (7000), 334

NH4 (1000), P (1000), K (3000), Ca (2000), SO4 (500), Mg (500), Cl (25), B (12.5), Mn 335

(1), Zn (1), Cu (0.25), Mo (0.25) and FeDTPA (100). For the low N treatment, NO3 and 336

NH4 were reduced to 70 and 10 µM, respectively, and K2SO4 was used to replace K and 337

SO4. Each mesocosm received two seeds and after 4 days they were thinned to one plant 338

per mesocosm. Plants were watered with 75 ml of deionized water every 2 days. Soil 339

solutions were collected at 20 cm depth intervals weekly using a micro-sampler 2.5 mm 340

in diameter and 9 cm in length (Soilmoisture Equipment CORP., Santa Barbara, CA, 341

USA). The solutions were stored at -80 oC until processing. The concentrations of nitrate 342

in the solutions were determined using the vanadium (III) chloride protocol according to 343

Doane et al. (2003). 344

Root harvest 345

The plants were harvested at 28 days after planting. At harvest a polyethylene liner in 346

each mesocosm was carefully removed and placed on a root washing station. The liners 347

were divided into 20 cm segments starting from the base of the shoot. Media were 348

carefully removed and the deepest layer reached by the roots was recorded for primary, 349

seminal, and crown root classes. CN in each nodal whorl and root branching were 350

counted. The roots were cut, separated from each segment, and preserved in 75% EtOH. 351

Total root lengths were obtained by scanning and analyzing using the WinRhizo software 352

(WinRhizo Pro, Régent Instruments, Québec City, Québec, Canada). 353

Shoot dry weight and plant nitrogen status 354

One day prior to harvest, leaf gas exchange of the first and the second youngest fully 355

expanded leaves was measured with a Licor-6400 Infrared Gas Analyzer (Li-Cor 356

Biosciences, Lincoln, NE, USA) using a red-blue light at PAR intensity of 1200 µmol 357

photons m-2 s-1 and constant CO2 concentration of 400 ppm. Shoot carbon assimilation 358

was measured with a Licor-6200 Infrared Gas Analyzer (Li-Cor Environmental Inc, 359






16

Lincoln, NE, USA). In short, a 36.5 liter (28x 28 x 46.5 cm) transparent acrylic chamber 360

was placed around a plant. The base of the chamber was split to fit a stem of a plant. The 361

air space around the stem and the base of the chamber was filled with modeling clay and 362

sponges to separate the shoot from the growth media. The chamber connected to the Li-363

6200 with polyethylene tubing 0.03 liter in volume. Carbon dioxide exchange was 364

measured for two minutes for each plant. Shoots were dried at 60 oC for 72h prior to dry 365

weight determination. The shoots were ground and 2 to 3 mg of ground tissue was taken 366

for tissue nitrogen analysis using an elemental analyzer (SeriesII CHNS/O Analyzer 367

2400, PerkinElmer). 368

Field studies 369

Field conditions, experimental design, and plant materials 370

Experiments were carried out during February to April in 2011 (SA2011) and 2012 371

(SA2012) at Alma, Limpopo province, South Africa (24°33′ 00.12 S, 28°07′25.84 E, 372

1235 masl) and during June - October in 2011 (US2011) at the Hancock Agricultural 373

research station of the University of Wisconsin in Hancock, WI, USA (44°07′56′′.74 N, 374

89°30′43′′.96 W, 331 masl). The soils at the experimental sites were a Clovelly loamy 375

sand (Typic Ustipsamment) in Alma and a Plainfield loamy sand (mixed, mesic Typic 376

Udipsamment) in Hancock. In SA2011 and SA2012 N fertilizers were applied at the rate 377

of 30 kg N/ha for 5 times until flowering resulting in 150 kg N ha-1 in total for well-378

fertilized plots. The low N plots received 30 kg N ha-1only at the beginning of the 379

growing season. In US2011 the well-fertilized plots were amended with 103 kg N ha-1 at 380

planting and at four weeks after planting resulting in a total of 206 kg N ha-1 while the 381

low N plots were amended with 34 kg N ha-1 at the beginning of the cropping season 382

only. In all environments, soil nutrient levels of other macro- and micronutrients were 383

adjusted to meet the requirements for maize production as determined by soil tests. Pest 384

control and irrigation were carried out as needed. 385

Plant material 386

The same six RILs used in the greenhouse experiment were used in SA2011. Different 387

sets of genotypes were planted at US2011 and SA2012. These genotypes were selected 388






17

based on previous screening in US field (Saengwilai et al., unpublished). Seven RILs 389

consisting of IBM1, IBM9, IBM13, IBM77, IBM133, IBM165, and IBM187 were used 390

in SA2012. RILs from the IBM populations; IBM10, IBM85, IBM218 and from the cross 391

between NY821 and H99 (NyH) population; NYH76, NYH57, NYH212 were used in 392

US2011. In each location the experiment was arranged in a split-plot design replicated 393

four times with high and low N treatments. Four sections adjacent to each other in the 394

field containing both high and low N treatments were assigned as blocks. Genotypes were 395

randomly assigned to five-row plots. Each row was 4.5 m long. The distance between 396

rows was 75 cm and within a row was 23 cm, resulting in a planting density of 6 plants 397

m-2. The plants were harvested at flowering, 9 weeks after planting in SA2011 and 398

SA2012 and 8 weeks after planting in US2011. 399

Root harvest 400

Evaluation of crown roots was carried out based on shovelomics (Trachsel et al., 2011). 401

Three representative plants were selected for excavation in each plot. The selection was 402

based on height, presence of bordering plants, and general appearance that represented 403

individuals in the plot. At harvest roots were excavated using spades. A large portion of 404

soil was removed from roots by carefully shaking. The remaining soil was removed by 405

soaking the roots in diluted commercial detergent followed by vigorously rinsing at low 406

pressure with water. Because three representative roots within a plot usually appear to be 407

homogeneous, only one root was selected for phenotyping. Crown root number (CN) was 408

measured by counting half of the root system. Assuming that the maize root system is 409

symmetrical, CN was multiplied by two to obtain the total CN prior to data analysis. Data 410

on other root traits such as root angle, diameter, and branching were also collected and 411

included in the analyses when needed. 412

Rooting depth and 15N injection 413

Rooting depth was measured at flowering by soil coring (Giddings Machine Co., 414

Windsor, CO, USA). Soil cores were taken within a planting row midway between two 415

plants. The diameter of soil cores was 5.1 cm. The cores were divided into 10 cm 416

segments and roots were extracted from each soil segment. Root lengths were obtained 417

by scanning and analysis using WinRhizoPro (Régent Instruments, Québec, Québec City 418






18

Canada). Percentages of root length at each depth were calculated in each soil core. 419

Depth above which 95% (D95) of root length is located was calculated by linear 420

interpolation between the cumulative root lengths (Trachsel et al.,2013). 421

The ability of roots to acquire N in deep soil layers was studied by deep injection of 422 15NO3

− in SA2012. PVC pipes with a length of 75 cm and a diameter of 5 cm were used 423

for 15NO3− injection. Three representative plants were selected and the injections were 424

done at a midway between adjacent plants within a planting row. Each plot received two 425

injections. Prior to the injections, a soil auger was used to excavate a cylinder of soil to a 426

depth of 50 cm. A PVC pipe was inserted into the hole and the 15NO3− solution was 427

poured into the hole. Each plot had 5 mL of K15NO3− solution (0.46 mg 15N mL-1, 98% 428

15N enriched) injected into each of two holes. Following the injection each hole was filled 429

with sand to prevent roots from growing down the hole. Seven days after 15NO3− 430

injection, the shoot biomass of the selected plant was harvested for 15N and total N 431

analysis. 432

Shoot dry weight and tissue nitrogen content 433

In SA2011 and SA2012 one day prior to harvest, leaf gas exchange of the ear leaves was 434

measured with a Licor-6400 Infrared Gas Analyzer (Li-Cor Biosciences, Lincoln, NE, 435

USA) using a red-blue light at PAR intensity of 1800 µmol photons m-2 s-1 and constant 436

CO2 concentration of 360 ppm. In all experiments, shoots were dried at 60 oC prior to dry 437

weight determination. The leaves and stems were ground and 2-3 mg of ground tissue 438

were analyzed for tissue nitrogen content using an elemental analyzer (SeriesII CHNS/O 439

Analyzer 2400, PerkinElmer).15N in plant tissue was analyzed using a PDZ Europa 440

ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass 441

spectrometer (Sercon Ltd., Cheshire, UK) at the Stable Isotope Facility, University of 442

California at Davis, USA (http://stableisotopefacility.ucdavis.edu/). 443

Statistical analysis 444

Statistical analyses were performed using R version 2.15.1 (R Development Core Team 445

2012). Linear mixed effect models were fit using the function lme from the package nlme 446

(Pinheiro et al., 2012) and two-way ANOVA were used for comparisons between high 447






19

and low CN groups (or individual genotypes), nitrogen levels and the interaction between 448

these main effects. ANCOVA was performed using the lm function to test effects of CN 449

and N treatments on response variables. The protected least significant difference post 450

hoc (α=0.05) test and Tukey’s Honest Significant Difference method (α=0.05) were used 451

for multiple comparison tests. 452

Acknowledgements 453

We thank Dr. Kathleen M. Brown for her helpful review of the manuscript, and Bob 454

Snyder, Bill Kojis, Curtis Frederick, and Johan Prinsloo for the management of the 455

experiments in Hancock and Alma. 456






20

Literature Cited 457

Azeez J, Adetunji M, Lagoke S (2006) Response of low-nitrogen tolerant maize 458 genotypes to nitrogen application in a tropical Alfisol in northern Nigeria. Soil 459 Tillage Res 91: 181–185 460

Bayuelo-Jiménez JS, Gallardo-Valdéz M, Pérez-Decelis VA, Magdaleno-Armas L, 461 Ochoa I, Lynch JP (2011) Genotypic variation for root traits of maize (Zea mays 462 L.) from the Purhepecha Plateau under contrasting phosphorus availability. Field 463 Crop Res 121: 350–362 464

Burton AL (2010) Phenotypic evaluation and genetic basis of anatomical and 465 architectural root traits in the genus Zea. Pennsylevania State University 466

Burton AL, Brown KM, Lynch JP (2013) Phenotypic diversity of root anatomical and 467 architectural traits in Zea species. Crop Sci 53: 1042–1055 468

Cassman KG, Doberman A, Walters DT (2002) Agroecosystems, nitrogen-use 469 efficiency, and nitrogen management. J Hum Evol 31: 132–140 470

Coudert Y, Périn C, Courtois B, Khong NG, Gantet P (2010) Genetic control of root 471 development in rice, the model cereal. Trends Plant Sci 15: 219–226 472

De Dorlodot S, Forster B, Pagès L, Price A, Tuberosa R, Draye X (2007) Root system 473 architecture: opportunities and constraints for genetic improvement of crops. Trends 474 Plant Sci 12: 474–481 475

Evans JR (1983) Nitrogen and photosynthesis in the flag leaf of wheat ( Triticum 476 aestivum L .). Plant Physiol 72: 297–302 477

Gastal F, Lemaire G (2002) N uptake and distribution in crops: an agronomical and 478 ecophysiological perspective. J Exp Bot 53: 789–799 479

Gaudin ACM, McClymont SA, Holmes BM, Lyons E, Raizada MN (2011) Novel 480 temporal, fine-scale and growth variation phenotypes in roots of adult-stage maize 481 (Zea mays L.) in response to low nitrogen stress. Plant Cell Environ 34: 2122–2137 482

Hetz W, Hochholdinger F, Schwall M, Feix G (1996) Isolation and characterization of 483 rtcs, a maize mutant deficient in the formation of nodal roots. Plant J 10: 845–857 484

Hochholdinger F (2009) The maize root system  : Morphology , Anatomy , and Genetics. 485 Handb. maize Its Biol. Springer Science, pp 145–160 486

Hochholdinger F, Woll K, Sauer M, Dembinsky D (2004) Genetic dissection of root 487 formation in maize (Zea mays) reveals root-type specific developmental 488 programmes. Ann Bot 93: 359–368 489






21

Hoppe DC, Mccully ME, Wenzel CL (1986) The nodal roots of Zea  : their development 490 in relation to structural features of the stem. Cannadian J Bot 64: 2524–2537 491

Jenkins MT (1930) Heritable characters of maize XXXIV-Rootless. J Hered 21: 79–80 492

Krassovsky I (1926) Physiological activity of the seminal and nodal roots of crop plants. 493 Soil Sci 21: 307–325 494

Kristensen HL, Thorup-Kristensen K (2004a) Uptake of 15N labeled nitrate by root 495 systems of sweet corn , carrot and white cabbage from 0. 2 – 2. 5 meters depth. Plant 496 Soil 265: 93–100 497

Kristensen HL, Thorup-Kristensen K (2004b) Root growth and nitrate uptake of three 498 different catch crops in deep soil layers. Soil Sci Soc Am J 68: 529–537 499

Lynch JP (2013) Steep, cheap and deep: an ideotype to optimize water and N acquisition 500 by maize root systems. Ann Bot 112: 347–357 501

Lynch JP (2007) Roots of the second green revolution. Aust J Bot 55: 493–512 502

Lynch JP, Ho MD (2005) Rhizoeconomics: carbon costs of phosphorus acquisition. 503 Plant Soil 269: 45–56 504

Mi G, Chen F, Zhang F (2005) Physiological and genetic mechanisms for nitrogen-use 505 efficiency in maize. J Crop Sci Biotechnol 10: 57–63 506

Miller AJ, Cramer MD (2004) Root nitrogen acquisition and assimilation. Plant Soil 507 274: 1–36 508

Nord E a, Lynch JP (2009) Plant phenology: a critical controller of soil resource 509 acquisition. J Exp Bot 60: 1927–1937 510

Pinheiro L, Bates D, DebRoy S, Sarkar D (2012) The nlme package; linear and 511 nonlinear mixed effects models. 216–225 512

Poudel D., Horwath W., Mitchell J., Temple S. (2001) Impacts of cropping systems on 513 soil nitrogen storage and loss. Agric Syst 68: 253–268 514

Raun WR, Johnson G V. (1999) Improving nitrogen use efficiency for cereal 515 production. Agron J 91: 357–363 516

Ribaudo M, Delgado J, Hansen L, Livingston M, Mosheim R, Williamson J (2011) 517 Nitrogen in agricultural systems: Implications for conservation policy. 518

Rubio G, Lynch JP (2007) Compensation among root classes in Phaseolus vulgaris L. 519 Plant Soil 290: 307–321 520






22

Taramino G, Sauer M, Stauffer JL, Multani D, Niu X, Sakai H, Hochholdinger F 521 (2007) The maize (Zea mays L.) RTCS gene encodes a LOB domain protein that is a 522 key regulator of embryonic seminal and post-embryonic shoot-borne root initiation. 523 Plant J 50: 649–59 524

Tian Q, Chen F, Liu J, Zhang F, Mi G (2008) Inhibition of maize root growth by high 525 nitrate supply is correlated with reduced IAA levels in roots. J Plant Physiol 165: 526 942–51 527

Trachsel S, Kaeppler SM, Brown KM, Lynch JP (2013) Maize root growth angles 528 become steeper under low N conditions. Field Crop Res 140: 18–31 529

Trachsel S, Kaeppler SM, Brown KM, Lynch JP (2011) Shovelomics  : high 530 throughput phenotyping of maize ( Zea mays L .) root architecture in the field. Plant 531 Soil 314: 75–87 532

Walk TC, Jaramillo R, Lynch JP (2006) Architectural Tradeoffs between Adventitious 533 and Basal Roots for Phosphorus Acquisition. Plant Soil 279: 347–366 534

Worku M, Bänziger M, Erley GS auf’m, Friesen D, Diallo AO, Horst WJ (2007) 535 Nitrogen uptake and utilization in contrasting nitrogen efficient tropical maize 536 hybrids. Crop Sci 47: 519–528 537

York LM, Nord EA, Lynch JP (2013) Integration of root phenes for soil resource 538 acquisition. Front Plant Sci 4: 1–15 539

Zhu J, Kaeppler SM, Lynch JP (2005) Mapping of QTL controlling root hair length in 540 maize ( Zea mays L .) under phosphorus deficiency. Plant Soil 270: 299–310 541

Zhu J, Mickelson SM, Kaeppler SM, Lynch JP (2006) Detection of quantitative trait 542 loci for seminal root traits in maize (Zea mays L.) seedlings grown under differential 543 phosphorus levels. Theor Appl Genet 113: 1–10 544

545

546






23

Figure legends 547

Figure 1. Visualization of maize root system of low and high crown root (CN) genotypes 548

at 40 d after germination. Crown roots are colored in blue and seminal roots are in red. 549

The number CN is 8 in the low CN genotypes and 46 in the high CN genotype (image 550

courtesy of Larry M. York). 551

Figure 2. Crown root number of maize 28 days after planting under high N and low N 552

conditions in soil mesocosms. Data shown are means of 4 replicates ± SE of the mean. 553

Means with the same letters are not significantly different (p < 0.05) 554

Figure 3. Crown root whorl number (3A) and crown root number per whorl (3B) of maize 555

28 days after planting under high N and low N conditions in soil mesocosms. Data shown 556

are means of six genotypes (i.e. IBM133, IBM123, OHW3, OHW61, OHW74, and 557

OHW170) with 4 replicates ± SE of the means. Means with the same letters are not 558

significantly different (p < 0.05) 559

Figure 4. Crown root number of maize at flowering under high N and low N conditions at 560

the fields in USA in 2011 (4A), and in South Africa in 2011 (4B) and 2012 (4C). Data 561

shown are means with 4 replicates ± SE of the means. Means with the same letters are not 562

significantly different (p < 0.05) 563

Figure 5 Rooting depth of six RILs at 28 DAP in soil mesocosms (5A), depth of primary, 564

seminal, and crown roots at 28 DAP under low N conditions in soil mesocosms compared 565

between high and low CN within the same root class (5B) and D95 of maize at 9WAP 566

under low and high N conditions at SA2011 field (5C). Data shown are means of 4 567

replicates + SE of the mean. Different letters represent significant differences (p<0.05). 568

Figure 6 Correlations between 6A) crown root number and rooting depth (R2=0.53, 569

p=0.04), 6B) 15N in shoot (R2=0.35, p=0.02), and 6C) shoot dry weight (R2=0.16, p=0.02) 570

at flowering under low N conditions in the field in South Africa (2012). 571

Figure 7 Correlation between crown root number and grain yield (% of yield under high 572

N) (R2=0.19, p=0.02) under low N conditions in the field in the USA. 573

Supplemental Figure S1 Crown root number per first (S1A), second (S1B), and third 574

(S1C) whorl of maize 28 days after planting under high N and low N conditions in soil 575

mesocosms. Data shown are means with 4 replicates ± SE of the means. Means with the 576






24

same letters are not significantly different (p < 0.05) 577

Supplemental Figure S2 Correlations between crown root number and rooting depth 578

(R2=0.68, p=0.04) at flowering under low N conditions in the field in USA. 579

580






25

Table I. Summary of correlation analysis (correlation coefficient and significant levels) 581

between crown root number and parameter measured under low N conditions in six 582

maize genotypes in soil mesocosms at 28 days after planting and in the field in South 583

Africa and USA in 2011. 584

585

Mesocosms Field

Parameter South Africa USA Canopy photosynthetic rate 0.26* - - Leaf photosynthetic rate 0.34* 0.31* - Tissue nitrogen content 0.23* 0.13* 0.13* Shoot dry weight 0.23* 0.49** 0.22*

*p<0.05, **p<0.01 586






26

Table II. Summary of ANCOVA model (F-value and degrees of freedom) of shoot traits 587

at 28 day after planting as influenced by CN and N treatment in six maize RILs in 588

greenhouse mesocosms. 589

Effect Shoot weight

Photosynthesis Rate

Carbon Assimilation

Tissue N Content

CN 22.31 (1,43)***

9.48 (1,43)**

6.51 (1,43)*

16.55 (1,44)*** N treatment 23.78 (1,43)***

31.62 (1,43)***

75.66 (1,43)***

29.15 (1,44)***

CN:N treatment 4.89 (1,43)*

14.08 (1,43)***

0.20 (1,43)

2.79 (1,44) a R2 0.66

0.53

0.63

0.49

†p<0.1, *p<0.05, **p<0.01, ***p<0.001, ap=0.10.Degrees of freedom shown as (numerator, denominator) 590






27

Table III. Summary of ANCOVA model (F-value and degrees of freedom) of shoot traits 591

at flowering as influenced by CN and N treatment in six maize RILs in SA2011 and 592

US2011 and in seven maize RILs in SA2012. 593

594

Effect Shoot weight

SA2011

Shoot weight SA2012

Shoot weight US2011

Yield US2011

CN 3.19 (1,44) †

0.89 (1,52)

0.84 (1,44)

21.37 (1,44)*** N treatment 63.28 (1,44)***

33.53 (1,52)***

22.39 (1,44)***

14.34 (1,44)***

CN:N treatment 1.10 (1,44)

3.05 (1,52) †

3.62 (1,44) †

2.67 (1,44) a R2 0.59

0.39

0.34

0.49

†p<0.1, *p<0.05, **p<0.01, **p<0.001,ap=0.10. Degrees of freedom shown as (numerator, denominator) 595

596






28

597

598

Figure 1. Visualization of maize root system of low and high crown root (CN) genotypes 599

at 40 d after germination. Crown roots are colored in blue and seminal roots are in red. 600

The number CN is 8 in the low CN genotypes and 46 in the high CN genotype (image 601

courtesy of Larry M. York). 602






29

Cro

wn

root

num

ber

0

2

4

6

8

10

12

14

bccd

a

bcb

de cd cdecd

f

bcd

ef

OHW170 OHW3 IBM133 IBM123 OHW61 OHW74

High CN Low CN High NLow N

603

604 605 Figure 2. Crown root number of maize 28 days after planting under high N and low N 606

conditions in soil mesocosms. Data shown are means of 4 replicates ± SE of the mean. 607

Means with the same letters are not significantly different (p < 0.05) 608

609






30

610

611 612 Figure 3. Crown root whorl number (3A) and crown root number per whorl (3B) of maize 613

28 days after planting under high N and low N conditions in soil mesocosms. Data shown 614

are means of six genotypes (i.e. IBM133, IBM123, OHW3, OHW61, OHW74, and 615

OHW170) with 4 replicates ± SE of the means. Means with the same letters are not 616

significantly different (p < 0.05). 617






31

010

3050

70

a

bcab

ccde cd cde cde

cd

decd

e

NYH76 NYH57 IBM10 IBM85 IBM218 NYH212

A High NLow N

Cro

wn

root

num

ber

010

2030

4050

aab ab

ab ababc

bcd bcd bcd

dbcd cd


B

010

2030

4050

60

bcd

aabc

ab

cdabc

cde cdeabcd

cde abcd

de de e

IBM165 IBM9 IBM187 IBM77 IBM13 IBM1 IBM133

C

Genotype 618

Figure 4. Crown root number of maize at flowering under high N and low N conditions in 619

the field in the USA in 2011 (4A), and in South Africa in 2011 (4B) and 2012 (4C). Data 620

shown are means with 4 replicates ± SE of the mean. Means with the same letters are not 621

significantly different (p < 0.05). 622






32

160

120

80

40

0A Primary Seminal Crown

a

b

a

b

a

b

60

50

40

30

20

10

0

a

aba

b

B High N Low N

Soi

l dep

th (c

m)

High CN Low CN

Soi

l dep

th (c

m)

160

140

120

100

80

60

40

20

0

bc

c

bcbc

abc abcab

a

ab

a

ab abc


High NLow NHigh CN Low CN

160

120

80

40

0A Primary Seminal Crown

a

b

a

b

a

b

60

50

40

30

20

10

0

a

aba

b

B High N Low N

Soi

l dep

th (c

m)

High CN Low CN

A

C

So

il d

ep

th (

cm

)

160

140

120

100

80

60

40

20

0

bc

c

bcbc

abc abcab

a

ab

a

ab abc


High NLow NHigh CN Low CN

B

Soi

l Dep

th (c

m)

623 Figure 5 Rooting depth of six RILs at 28 DAP in soil mesocosms (5A), depth of primary, 624

seminal, and crown roots at 28 DAP under low N conditions in soil mesocosms compared 625

between high and low CN within the same root class (5B) and D95 of maize at 9WAP 626

under low and high N conditions in the SA2011 field study (5C). Data shown are means 627

of 4 replicates + SE of the mean. Different letters represent significant differences 628

(p<0.05). 629






33

25 30 35 40 45 50

10

20

30

40

50

Soi

l dep

th (c

m) A

15 20 25 30 35 40 45 50

0.05

0.10

0.15

0.20

0.25

0.30

N15

in s

hoot

(mg plant−1 )

B

25 30 35 40 45 50 55 60

30

40

50

60

70

Crown root number

Sho

ot d

ry w

eigh

t (g)

C

630 631 Figure 6 Correlations between 6A) crown root number and rooting depth (R2=0.53, 632

p=0.04), 6B) 15N in shoot (R2=0.35, p=0.02), and 6C) shoot dry weight (R2=0.16, p=0.02) 633

at flowering under low N conditions in the field in South Africa (2012). 634






34

20 25 30 35 40 45 50

0

20

40

60

80

100

Crown root number

Gra

in y

ield

(%)

635 Figure 7 Correlation between crown root number and grain yield (% of yield under high 636

N) (R2=0.19, p=0.02) under low N conditions in the field in the USA. 637

638






1 running head: fewer crown roots improve n capture in

Documents